JP5261510B2 - Network monitoring apparatus, method and program - Google Patents

Description

  The present invention relates to a reliability evaluation technique in a communication network, and in particular, a floor structure network in which the number of devices and the number of component modules in one device change with time, and a plurality of types of devices are used on the same floor. The present invention relates to a network monitoring apparatus, method, and program for use in identifying a suspected failure location when a failure occurs.

  Generally, in network monitoring, SNMP (Simple Network Management Protocol) (see Non-Patent Documents 1 and 2, for example) is used, and the SNMP agent of the monitoring target device detects abnormality (TRAP) to the SNMP manager of the monitoring device. Failure is detected by sending. In addition, the monitoring target device is periodically monitored in combination with alive monitoring by ICMP (Internet Control Message Protocol) (PING).

  However, in the case of the network configuration as shown in FIG. 1, the SNMP agent sends an abnormal value detection (TRAP) to the monitoring device which is the SNMP manager in the data connection network (DCN) when the device A fails in the monitored network. If it is detected, it is confirmed by the maintenance person by log confirmation or status confirmation command, and if it is determined that replacement is necessary, there are two possible locations for failure, the interface of device A and the interface of device C. It is often difficult to further narrow down. In such a case, if this failure affects the service, both the devices A and C are arranged simultaneously in order to shorten the service interruption time as much as possible. At this time, since the suspicious failure location cannot be specified, the cost of rushing to the field and arranging the goods is doubled, resulting in an increase in OPEX (operation cost).

  To solve this problem, a technique for identifying a suspected place using an evaluation measure such as a failure rate has been proposed (see, for example, non-patent literature).

  In addition, there is a method for estimating lifetime parameters by using a maximum likelihood estimation method based on device data acquired by MIB (Management Information Base) information, obtaining a confidence interval, and evaluating a failure rate (for example, patents) Reference 1).

  Also, a monitoring system has been proposed that compares the pre-recorded device failure rate confidence interval with the failure rate based on the cumulative failure number of the network device to determine whether the network device is reliable or not (for example, Patent Document 2). reference).

JP 2004-191359 A JP 2007-68090 A

RFC1157 A Simple Network Management Protocol (SNMP) RFC1213 Management Information Base of Network Management of TCP / IP-based Internets: MIB-II. Hiroshi Shiomi, “Basics of Reliability”, “2.2 Calculation of Failure Rate”, pp20-23, Corona Inc., 1975.

  When the failure rate is used to identify the suspected failure location, the minimum unit for replacement is a module, not the entire device. Here, a module is defined as the smallest unit of parts that can be replaced independently. For this reason, the failure rate should be compared on a module-by-module basis to identify the suspected location.

  The failure rate for each module is obtained from Non-Patent Document 3 by dividing the number of failures by the total operating time (= measurement time x total number of modules), so if the information on the number of modules cannot be obtained accurately, Significantly affects the calculation and comparison of rates. For example, when the number of modules is excessively counted, the failure rate is estimated to be small, and the cause of the failure having a large problem is overlooked. In the opposite case, there is a possibility of overestimating the cause of the failure with a small problem.

  The same can be said for MTBF (Mean Time Between Failure), which is the reciprocal of the failure rate.

  In a communication network, network devices are added as the number of users changes over time. At this time, the communication carrier does not mount all the modules at the time of expansion of the device in consideration of CAPEX (= capital investment cost), but there is also a module in which expansion is performed in a form that predicts demand. For this reason, it can be said that the number of modules is changing with time.

  In recent years, networks have become multi-vendored, and devices of the same vendor in a network with a hierarchical structure are often used. In other words, it is difficult to calculate the total number of modules in all devices in a practical operation.

  When acquiring the number of modules using the management information (MIB information) of the device held by the SNMP agent, frequent information collection has the risk of increasing the load on the device, and there is also a concern that the monitoring traffic will increase. Conceivable. In recent years, as network devices have become multi-vendors, interface products have become more diversified, so there are cases where module information cannot be distinguished only by MIB information. From the above, it is not desirable to collect module information using general MIB information.

  Both of the technologies of the above-mentioned Patent Documents 1 and 2 are information in units of devices, and the failure rate in units of modules is not calculated. Even if the suspected device can be identified, the suspected module cannot be identified. When calculating the failure rate for each module, information on the total number of modules is required. However, since the number of devices and the number of modules change over time, and the order of the number of devices in the current communication network is more than tens of thousands, information on changes in the number of all modules It is difficult to collect data, and an estimation method is required when all module information cannot be collected.

  The present invention has been made in view of the above-described points, and in particular, a floor structure in operation in which a plurality of types of devices are used on the same floor, in which the number of devices and the number of component modules in one device change over time. An object of the present invention is to provide a network monitoring apparatus, method, and program capable of calculating a module failure rate by estimating the number of constituent modules in a network and using the result to identify a suspected location when a failure occurs. And

In order to solve the above-mentioned problems, the present invention changes the number of devices for a hierarchical structure network in which a plurality of device types exist on the same floor or the number of component modules in one device over time, so that there are a plurality of types on the same floor. In a hierarchical structure network where the device is used, a monitoring device that identifies a suspected failure location when a failure occurs,
Network failure information including the suspected failure module information,
Number of users time series information,
Device module configuration information including a flag that changes the maximum number of modules that can be mounted per device and the number of modules that constitute the device, the maximum number of users that can be accommodated for each module, and the device floor type,
Number of devices time series information,
Network configuration information, including interface counterparts and modules;
User number prediction parameter and capacity factor parameter used for user number prediction,
Storage means storing
Classifying means for classifying the network failure information by device / module, and calculating the number of failures for each module;
Change determination means for classifying the device module configuration information based on the flag as to whether or not the number of modules to be evaluated changes in the device;
Module number calculation for calculating the number of modules from the maximum number of modules installed per device in the device module configuration information and the number of devices in the device number time-series information when the change determining means determines that “no change” Means,
Using the device layer type in the device module configuration information, the number of users is distributed from the device number ratio of the same floor from the device information of the same floor device and the device number information of the device number time series information, and the number of users. A user number prediction means for obtaining the number of future users from the number of users classified as a prediction parameter;
From the accommodation rate parameter and the maximum number of users that can be accommodated for each module in the device module configuration information, module number prediction means for obtaining the number of modules;
When it is determined that the change determination means is “changed”, the number of modules obtained by the module number prediction means is obtained, and when it is determined that “no change”, it is obtained by the module number calculation means. Module failure rate calculating means for calculating the module failure rate for each module and storing the module failure rate for each module using the number of modules, and
A suspected failure point extraction means for extracting a suspected failure location candidate from the network configuration information and failure detection information detected by any means;
Output means for extracting and displaying a module failure rate of a corresponding suspected failure location with reference to the failure rate storage means based on the suspected failure location candidate.

  As described above, according to the present invention, the total number of modules is estimated particularly in a network in which the number of devices and the number of component modules in one device change with time, and multiple types of devices are used in the same level. By doing so, the failure rate for each module can be calculated. This makes it possible to prioritize the cause location when a failure occurs. As a result, it is possible to reduce useless maintenance operations that have become redundant, thereby reducing operational costs.

It is a figure which shows an example of the relationship between the network apparatus and monitoring apparatus in one embodiment of this invention. It is a block diagram of the monitoring apparatus in one embodiment of this invention. It is an example of the network failure information of failure management DB in one embodiment of this invention. It is an example of the user number time series information of the user management DB in an embodiment of the present invention. It is an example of the apparatus module structure information of equipment management DB in one embodiment of this invention. It is an example of the apparatus number time series information of equipment management DB in one embodiment of the present invention. It is an example of the network configuration information of equipment management DB in one embodiment of the present invention. It is a flowchart of the module failure rate calculation process in one embodiment of the present invention. It is an example of the user number prediction parameter and the accommodation rate parameter stored in the parameter memory | storage part in one embodiment of this invention. It is a flowchart of failure suspected place identification processing in an embodiment of the present invention. It is an example which classified the number of monthly failures for each module in one embodiment of the present invention. It is an example of module number calculation in case the number of modules does not change in one Example of this invention. It is an example which calculated the number of distribution users from the apparatus number ratio for every apparatus of the same floor in one Example of this invention. It is an example which estimated the number of distribution users by the linear function approximation of the least squares method in one Example of this invention. It is an example of module number calculation in case the number of modules which comprise the apparatus in one Example of this invention changes. It is an example of module failure rate calculation in one Example of this invention. It is a connection structural example in one Example of this invention.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 shows an example of the relationship between a network configuration and a monitoring device according to an embodiment of the present invention.

  As the entire network, there are four types, that is, a core network 20, an access network 30, a home network 40, and a data connection network (DCN) 10, and the monitoring target networks are the core network 20 and the access network 30, and the monitoring device 100 is It exists in the data connection network 10.

  Each network to be monitored is composed of network devices, each having a hierarchical structure. Assuming that the devices on the same floor have a plurality of device types.

  The monitoring device 100 and the monitoring target device use the SNMP protocol or the like to detect a failure. Specifically, the monitoring device 100 has an SNMP manager function 200, and the monitoring target device 21 has an SNMP agent function 22.

  The embodiment will be described on the premise of the above network configuration.

  FIG. 2 is a diagram showing a configuration of a monitoring device and a flow of processing thereof according to an embodiment of the present invention. The monitoring apparatus 100 shown in the figure includes an apparatus information receiving unit 111 that receives information from an external input unit 101, a user time series information receiving unit 112, a failure information receiving unit 113, a parameter receiving unit 114, an equipment management DB (database). 110, user management DB 120, failure management DB 130, failure rate management DB 140, module failure rate calculation unit 150, failure suspect candidate extraction unit 160, parameter storage unit 171, failure rate storage unit 172, extraction result storage unit 173, monitoring system 201 A network interface 202 and an SNMP manager 200.

  As shown in FIG. 3, the failure management DB 140 stores the failure device, the failure occurrence date and time, and the failure suspect module as the network failure information received by the failure information receiving unit 113.

  As shown in FIG. 4, the user management DB 120 stores the information acquisition date and the total number of users as the user number time-series information received by the user time-series information receiving unit 112.

  The equipment management DB 110 stores device module configuration information, device number time series information, and network configuration information received by the device information receiving unit 111. As the device module configuration information, as shown in FIG. 5, a device type, a device floor type, a module type, a maximum number of modules that can be mounted per device, a flag indicating whether the number of modules constituting the device can change, Includes the maximum number of users that can be accommodated for each module. Further, as shown in FIG. 6, the device number time series information includes the device type, the number of devices, and the information acquisition date. As shown in FIG. 7, the network configuration information includes a device type, an interface, a corresponding module, a counter device, a counter interface, and a counter module.

  Here, the “device floor type” is information that can determine a device group of the same floor. The target network below has a network structure as shown in FIG. 1, and has a network structure having a hierarchical structure as a whole. The network configuration information is information for distinguishing the same floor and different models because it is assumed that multi-vendor devices are accommodated for the same floor devices.

  The parameter receiving unit 114 performs the following process.

  FIG. 8 is a flowchart of module failure rate calculation processing according to an embodiment of the present invention.

  Step 110) It is assumed that the parameter receiving unit 114 acquires the user number prediction parameter and the accommodation rate parameter from the input unit 101 and stores them in the parameter storage unit 171 as shown in FIG. The module failure rate calculation unit 150 receives the network failure information from the failure management DB 130, the user number time series information from the user DB 120, the device module configuration information and the device time series information from the facility management DB 110, and the user shown in FIG. 9 from the parameter storage unit 171. The number prediction parameter and the accommodation rate parameter are extracted.

  Step 120) The module failure rate calculation unit 150 classifies the network failure information extracted from the failure management DB 130 by device and by failure suspect module, and calculates the number of failures for each module for a specific period.

  Step 130) The module failure rate calculation unit 150 refers to a flag indicating whether or not the number of modules constituting the device changes, and when it changes (flag = “1”), the module failure rate calculation unit 150 proceeds to Step 150 and does not change. If the flag is “0”, the process proceeds to step 140.

  Step 140) When the number of modules constituting the device does not change, the time series change of the number of devices and the change in the number of modules follow, so the number of modules can be calculated using information on the maximum number of modules mounted per device. It becomes possible. That is, the module failure rate calculation unit 150 calculates the number of modules from the maximum number of modules mounted per device in the device module configuration information and the number of devices in the device number time-series information, and stores them in the failure rate storage unit 172. Control goes to step 170.

  Step 150) When the number of modules constituting the apparatus can change, the module failure rate calculation unit 150 predicts the number of users when predicting the number of modules. This is based on the assumption that module / equipment addition determination at the design stage is made in proportion to the increase in the number of users. The ratio of the number of devices in the same device is obtained using the device floor type information, and the accommodation user information for each device is calculated. Hereinafter, the number of users distributed according to the ratio of the number of devices is defined as the number of distributed users. The number of allocated users in the future is predicted by the least square method using the input user number prediction parameter.

  Step 160) Further, the module failure rate calculation unit 150 estimates the number of modules from the input accommodation rate parameter, the maximum number of users that can be accommodated for each module, and the number of distributed users predicted in Step 150, and stores the value as the failure rate storage. Stored in the unit 172.

  Step 170) The module failure rate calculation unit 150 calculates and estimates at the above-described Step 140 and Step 160, and stores the number of modules stored in the failure rate storage unit 172, the number of failures for each module in a certain period, and the period information. From the above, the failure rate is calculated by the method shown in Non-Patent Document 3 described above.

  Step 180) The calculated module failure rate is stored in the failure rate DB 140.

  The above processing is performed for all devices and all modules.

  FIG. 10 is a flowchart of the suspected fault location specifying process according to the embodiment of the present invention.

  Step 210) The suspected failure candidate extraction unit 160 acquires TRAP information that the monitoring system 201 has detected a failure due to the SNMP TRAP function or the like.

  Step 220) The suspected failure candidate extraction unit 160 extracts suspected failure location candidates based on the received TRAP. Information. At this time, if a failure occurs in an interface or the like, the module to which the interface is connected is also considered as a suspicious object. Therefore, network configuration information (apparatus type, interface, module type, connection counter device, connection counter interface, module type) (FIG. 7) (FIG. 7) is extracted from the facility management DB 110, and modules of adjacent devices are also subject to suspicion.

  Step 230) The module failure rate of the suspected failure location candidate is extracted from the failure rate management DB 140 and displayed on the display unit 102.

  The above embodiment will be specifically described below.

  Hereinafter, processing of the module failure rate calculation unit 150 and the failure suspected candidate extraction unit 160 will be described. As the handling of time-series data in the present embodiment, when the information input accuracy in a specific period is described, it is assumed that this period can be arbitrarily determined. In each of the diagrams showing examples, the description is made on the assumption that data for each month and information at the end of the month are input.

  First, network failure information, number-of-users time-series information, device module configuration information, number-of-devices time-series information, network configuration information, number-of-users prediction parameters, and capacity ratio parameters are input from the input unit 101.

  In the present embodiment, the above information is received by the administrator via the receiving units 111 to 113, the network failure information is stored in the failure management DB 130, the user number time series information is stored in the user management DB 120, the device module configuration information, the number of devices. Series information and network configuration information are stored in the facility management DB 110. Similarly, the user number prediction parameter and the accommodation rate parameter are also stored in the parameter storage unit 171 via the parameter reception unit 114.

  As shown in FIG. 3, the network failure information of the failure management DB 140 includes information on the failure device, the failure occurrence date and time, and the suspected failure module.

  The user time series information in the user management DB 120 includes information on the total number of users accommodated in the monitoring target network, which is aggregated every specific period. FIG. 4 is an example of the monthly user count time series information.

  In the device module personality information of the facility management DB 130, as shown in FIG. 5, the device type, the device floor type, the module type, the maximum number of modules that can be mounted per device, and the number of modules constituting the device can change. Flag and the maximum number of users that can be accommodated for each module.

  The number-of-devices time series information in the facility management DB 130 includes information on the device type, the number of devices, and information acquisition date as shown in FIG.

  As shown in FIG. 7, the network configuration information of the facility management DB 130 includes a device type, an interface, a module type, a connected device, a connected device, and a module type.

  As shown in FIG. 9, the user number prediction parameter stored in the parameter storage unit 171 includes a past data period used for prediction and a future period to be predicted, and the accommodation rate parameter includes end rate information. Is entered by the administrator.

  The operation of the present embodiment will be described below with reference to FIG.

  Network failure information, number-of-users time-series information, device module configuration information, number-of-devices time-series information, network configuration information, number-of-users prediction parameter, and capacity factor parameter are input. This procedure is a procedure performed by the administrator in advance, and is performed separately from the following processing.

  Step 110) The module failure rate calculation unit 150 extracts information stored in advance from each DB.

  Step 120) The module failure rate calculation unit 150 acquires the device type and the module type from the device module configuration information, and classifies the failure for each of the suspected module information in the network failure information. The number of failure occurrences is totaled for each specific period and stored in the failure rate storage unit 172. FIG. 11 shows the result of totaling the number of failures of each module in the devices A and C obtained by month.

Step 140) <Calculating the number of modules when the number of modules in the apparatus does not change>
The module failure rate calculation unit 150 uses the flag information as to whether the number of modules constituting the device in the device module configuration information can change or not. Perform the calculation. In this case, since the number of devices follows the number of modules, that is, the maximum number is always installed, the maximum number of modules that can be mounted per device in the device module information and the number of devices in the time series information The number of modules is calculated by multiplying by and stored in the failure rate storage unit 172. FIG. 12 is an example in which the number of modules in which the number of modules constituting the apparatus in “July 2010” does not change is calculated from the above-described information in FIGS. 5 and 6.

Step 150) <Prediction of the number of modules when the number of modules in the apparatus changes>
The module failure rate calculation unit 150 predicts the number of users for a module that is flagged as “change” (= 1) using flag information indicating whether the number of modules constituting the device in the device module configuration information can change. The number of users for the predicted future period in the parameter is predicted, and the number of modules necessary to accommodate the user is obtained. Further, it is assumed that the target network is composed of a plurality of types of devices for the same floor. That is, it is assumed that the total number of users is distributed according to the ratio of the number of devices on the same floor.

  In the method for predicting the number of users, first, as described above, the ratio of the number of devices is obtained from the number-of-devices time-series information for devices having the same device floor type in the device module configuration information. It is determined that this ratio is the user ratio between the devices, and the number of distributed users is calculated. FIG. 13 shows an example of the number of distributed users. From the device floor type information in the device module configuration information shown in FIG. 5, it can be seen that device A and device B are devices of the same floor. The number of distributed users is calculated from the number of users in the user series in FIG. 4 using the ratio of the number of apparatuses A and B in the apparatus number time-series information in FIG.

  For the prediction of the number of users, a total of three parameters are used: two parameters in the user number prediction parameter stored in the parameter storage unit 171 and one parameter in the accommodation rate parameter. The past data period used for the prediction is A, the future period to be predicted is B, and the number of modules is obtained at the time series N. Then, the number of distributed users in time series N + B (future) is calculated from the information on the number of distributed users in time series N-A + 1 to N using a least-squares linear function approximation.

Let U _t be the number of distributed users in time series t month. When obtaining the number of N modules, the difference up to time series N in time series t is defined as M _t . That is, M _N = 0. (M _{N-A + 1} , U _{N-A + 1} ) ... (0, U _N ) Time series N-A + 1 to N to minimize the change in the number of users using time series difference and distributed user number data A linear function approximation (y = ax + b) is performed by the square method, and parameters (slope a and intercept b) are calculated. Here, as described above, x is the time series difference M _t up to the time series N, and y is the number of classified users U _t .

上記で求めたaとｂの情報を元に、時系列N+B(未来)の分配ユーザ数を計算する。

Based on the information of a and b obtained above, the number of distributed users in time series N + B (future) is calculated.

  FIG. 14 shows the number of classified users for B = 6 months (January 2011) based on the information for A = 3 months from the data of the number of classified users in FIG. Is an example of predicting.

Step 160) <Estimating the number of modules when the number of modules in the apparatus changes>
The module failure rate calculation unit 150 uses the upper limit number of users that can be accommodated for each module and the accommodation rate parameter C% in the device module configuration information to determine the time for the time-series N + B (future) distribution users obtained in step 150. Estimate the number of modules in series N. That is, it is considered that the number of modules in time series N is determined so that C% of the maximum number of users that can be accommodated can be accommodated in anticipation of the number of distributed users at time series N + B (future).

  Specifically, the time-series N + B (future) number of distributed users obtained in step 150 is calculated by dividing the number of users that can be accommodated by each module by the capacity limit parameter, and the failure rate storage. Stored in the unit 172. FIG. 15 is an example of the number of modules obtained by the above method.

Step 170) <Calculation of module failure rate>
As described in Non-Patent Document 3, the module failure rate calculation unit 150 obtains the number of failures for each module classified in step 120 in steps 140 and 160, and stores the number of modules stored in the failure rate storage unit 172. The module failure rate is calculated by dividing by the time series difference. FIG. 16 is an example showing the module failure rate for each month in the apparatuses A and C when N = July 2010.

  Step 180) The module failure rate calculation unit 150 records the module failure rate calculated in Step 170 in the failure rate management DB 140.

  The above processing is repeated for the module types in all devices.

  Step 210) Unlike the processing up to step 180 described above, the failure suspected place identifying process in the failure suspect candidate extraction unit 160 is performed in a real-time monitoring operation.

  A failure is detected as TRAP for the monitored networks 20 and 30 using SNMP or the like. In this case, detection by ICMP (Ping) may be used.

Step 220) <Extracting a suspected fault candidate>
The suspected failure candidate extraction unit 160 determines a suspected failure location candidate based on the received TRAP information. The suspected failure location candidates are basically all modules of the TRAP issuing device. In addition, in the case of an interface state change alarm, the module of the opposite device of the corresponding interface is also set as a suspected place candidate using the network configuration information. In the case of other alarms, all modules of the device that issued the TRAP are set as suspected place candidates.

  The above processing is performed automatically, but the interface of the adjacent device based on the network configuration information can also be manually set as a suspected place candidate.

  FIG. 17 shows a connection configuration example of the device A and the device C. Module C of device A and module a of device C are connected. Module b is a module including module a. In this case, when a state change TRAP related to the module a of the device C is detected, there are five types of failure suspected place candidates: all the modules a, b, c, d of the device C and the module C of the device A.

Step 230) <Display module failure rate of suspected failure point candidate>
The failure rate of the corresponding module is obtained from the failure rate management DB 140 for the suspected failure location candidate module, read into the extraction result storage unit 173, and displayed. Based on this result, the operator can take a priority action. In the examples of FIGS. 16 and 17, it can be determined that the module C in the device A has the highest priority for handling among the suspected failure point candidates.

  Note that the processing and functions performed by each component of the monitoring device shown in FIG. 2 are the same as the processing performed by each unit using hardware resources such as a CPU and memory built in a computer used as the monitoring device. This is realized by executing the corresponding program. The program can be distributed to the market via a recording medium such as a flexible disk, a CD-ROM, a DVD, or a network such as the Internet.

  The present invention is not limited to the above-described embodiments and examples, and various modifications and applications are possible within the scope of the claims.

10 Data connection network (DCN)
20 Core Network 21 Device 22 SNMP Agent 30 Access Network 40 Home Network 100 Monitoring Device 101 Input Unit 102 Display Unit 111 Device Information Receiving Unit 112 User Time Series Information Receiving Unit 113 Fault Information Receiving Unit 110 Facility Management DB
120 User management DB
130 Failure management DB
140 Failure rate management DB
150 module failure rate calculation unit 160 failure suspected candidate extraction unit 171 parameter storage unit 172 failure rate storage unit 200 SNMP manager 201 monitoring system 202 network interface

Claims (7)

In a floor structure network in which the number of devices for a floor structure network in which a plurality of device types exist on the same floor and the number of component modules in one device change over time, and a plurality of types of devices are used on the same floor, A monitoring device that identifies a suspected failure location when a failure occurs,
Network failure information including the suspected failure module information,
Number of users time series information,
Device module configuration information including a flag that changes the maximum number of modules that can be mounted per device and the number of modules that constitute the device, the maximum number of users that can be accommodated for each module, and the device floor type,
Number of devices time series information,
Network configuration information, including interface counterparts and modules;
User number prediction parameter and capacity factor parameter used for user number prediction,
Storage means storing
Classifying means for classifying the network failure information by device / module, and calculating the number of failures for each module;
Change determination means for classifying the device module configuration information based on the flag as to whether or not the number of modules to be evaluated changes in the device;
Module number calculation for calculating the number of modules from the maximum number of modules installed per device in the device module configuration information and the number of devices in the device number time-series information when the change determining means determines that “no change” Means,
Using the device layer type in the device module configuration information, the number of users is distributed from the device number ratio of the same floor from the device information of the same floor device and the device number information of the device number time series information, and the number of users. A user number prediction means for obtaining the number of future users from the number of users classified as a prediction parameter;
From the accommodation rate parameter and the maximum number of users that can be accommodated for each module in the device module configuration information, module number prediction means for obtaining the number of modules;
When it is determined that the change determination means is “changed”, the number of modules obtained by the module number prediction means is obtained, and when it is determined that “no change”, it is obtained by the module number calculation means. Module failure rate calculating means for calculating the module failure rate for each module and storing the module failure rate for each module using the number of modules, and
A suspected failure point extraction means for extracting a suspected failure location candidate from the network configuration information and failure detection information detected by any means;
Based on the suspected failure location candidate, referring to the failure rate storage means, the module failure rate of the corresponding suspected failure location is extracted and displayed, and
A network monitoring apparatus comprising: The classification means includes
The network fault information, classified by device-specific modules, network monitoring device of claim 1 further comprising a means for calculating the fault count for each module for each specified period. When the change determination means determines that the number of modules to be evaluated is “changes” in the apparatus,
Before Symbol number of users predicting means,
Using the apparatus revised classification before Symbol module configuration information, from the device number ratio of the same revised determined from the information of the device number information and the device number time series information of the same revised apparatus distributes the number of users, the user predicted Including means for predicting the number of future users from the number of users classified as a parameter,
The module number predicting means includes:
The network monitoring device according to claim 1, further comprising means for predicting the number of modules from the accommodation rate parameter and the maximum number of users that can be accommodated for each module in the device module configuration information. In a floor structure network in which the number of devices for a floor structure network in which a plurality of device types exist on the same floor and the number of component modules in one device change over time, and a plurality of types of devices are used on the same floor, A network monitoring method for identifying a suspected failure location when a failure occurs,
Input means
Network failure information including the suspected failure module information,
Number of users time series information,
Device module configuration information including a flag that changes the maximum number of modules that can be mounted per device and the number of modules that constitute the device, the maximum number of users that can be accommodated for each module, and the device floor type,
Number of devices time series information,
Network configuration information, including interface counterparts and modules;
User number prediction parameter and capacity factor parameter used for user number prediction,
And an information acquisition step of storing in the storage means;
A classifying step for classifying the network failure information by device / module, and calculating the number of failures for each module;
A change determining unit that classifies the device module configuration information based on the flag as to whether or not the number of modules to be evaluated changes in the device;
When the module number calculation means determines that “no change” in the change determination step,
A module number calculating step of calculating a module number from the maximum number of modules mounted per device in the device module configuration information and the number of devices in the device number time-series information;
The number-of-users predicting means uses the number of devices in the device module configuration information to determine the number of users based on the device number ratio of the same floor from the information on the same floor device and the information on the number of devices in the device number time series information. A number-of-users prediction step of determining the number of future users from the number of users classified and classified as the number-of-users prediction parameter;
A module number predicting unit calculates a module number from the accommodation capacity parameter and the maximum number of users that can be accommodated for each module in the device module configuration information;
When the module failure rate calculation means is determined to be “changed” in the change determination step, the module number obtained in the module number prediction step is determined to be “not changed”. A module failure rate calculating step of calculating a module failure rate for each module using each of the modules obtained in the module number calculating step and storing the module failure rate in a failure rate storage unit;
The suspected failure point extraction means extracts the suspected failure location candidate from the network configuration information and failure detection information detected by any means, suspected failure location extraction step;
An output means for extracting and displaying a module failure rate of a corresponding suspected failure location with reference to the failure rate storage means based on the suspected failure location candidate; and
A network monitoring method characterized by: In the classification step,
Wherein the network fault information, classified by device-specific modules, network monitoring process according to claim 4, wherein for calculating the fault count for each module for each specified period. In the change determining step, when it is determined that the number of modules to be evaluated is “changed” in the apparatus,
In the previous Symbol number of users prediction step,
Using the apparatus revised classification before Symbol module configuration information, from the device number ratio of the same revised determined from the information of the device number information and the device number time series information of the same revised apparatus distributes the number of users, the user predicted Estimate the number of future users from the number of users classified as parameters,
In the module number prediction step,
The network monitoring method according to claim 5, wherein the number of modules is predicted from the accommodation rate parameter and the maximum number of users that can be accommodated for each module in the device module configuration information.   The network monitoring program for functioning a computer as each means which comprises the network monitoring apparatus of any one of Claim 1 thru | or 3.

Images